Abstract
Constraint solving is applied in different application contexts. Examples thereof are the configuration of complex products and services, the determination of production schedules, and the determination of recommendations in online sales scenarios. Constraint solvers apply, for example, search heuristics to assure adequate runtime performance and prediction quality. Several approaches have already been developed showing that machine learning (ML) can be used to optimize search processes in constraint solving. In this article, we provide an overview of the state of the art in applying ML approaches to constraint solving problems including constraint satisfaction, SAT solving, answer set programming (ASP) and applications thereof such as configuration, constraint-based recommendation, and model-based diagnosis. We compare and discuss the advantages and disadvantages of these approaches and point out relevant directions for future work.
Highlights
Over several decades, constraint solving (we use this as a general term referring to specific problem solving approaches such as constraint satisfaction (Freuder, 1997), Satisfiability Testing (SAT) solving (Gu et al, 1996), and answer set programming (Brewka et al, 2011)) has shown to be a core technology of Artificial Intelligence (Apt, 2003; Brailsford et al, 1999; Rossi et al, 2006)
A few approaches focus on aspects that go beyond basic performance analyses and classification accuracy (Erdeniz et al, 2019)
We have analyzed and summarized research contributions related to the integration of machine learning with constraint solving including (1) the basic approaches of constraint satisfaction, SAT solving, and answer set programming, and (2) applications thereof including configuration, diagnosis, and constraint-based recommendation
Summary
Constraint solving (we use this as a general term referring to specific problem solving approaches such as constraint satisfaction (Freuder, 1997), SAT solving (Gu et al, 1996), and answer set programming (Brewka et al, 2011)) has shown to be a core technology of Artificial Intelligence (Apt, 2003; Brailsford et al, 1999; Rossi et al, 2006). The n-queens problem is related to the task of placing n queens on n distinct squares of an n × n chessboard in such a way that no pair of queens attacks each other. The car sequencing problem is another problem that can be tackled on the basis of constraint solving techniques This problem is related to the task of scheduling a car production sequence in such a way that capacity constraints of the assembly line are satisfied. Each section has a limited capacity which has to be taken into account by the solver Another application area of constraint-based knowledge representations and reasoning is knowledge-based configuration (Felfernig et al, 2014). Constraint-based technologies are applied in a variety of further scenarios such as scheduling, vehicle routing, and recommender systems just to mention a few (Burke, 2000; Felfernig & Burke, 2008; Rossi et al, 2006)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
